Sensitivity Equalization and Dynamic Range Expansion with Multiple Optofluidic Microbubble Resonator Sensors

Abstract

A novel multi-optofluidic microbubble resonator (OMBR) sensitivity equalization method is presented that equalizes the sensing signal from different OMBRs. The method relies on the fact that the ratio of the wavelength shifts to the bulk refractive index sensitivity (BRIS) does not depend on the physical dimensions of the OMBR. The proof of concept is experimentally validated and the sensing signals from individual OMBRs can be directly compared. Furthermore, a wide dynamic range of sensing with favorable consistency and repeatability is achieved by piecing together signals from 20 OMBRs for HIV-1 p24 antigen detection from 50 fg/mL to 100 ng/mL (2.1 fM to 4.2 nM), indicating significant potential for practical applications, such as in drug screening and disease diagnosis.

Keywords: HIV-1 p24; WGM; biosensing; optofluidic.

Figure 1

Experimental setup of OMBR sensing system, including optical and microfluidic paths.

Figure 2

(a) Tapered fiber coupled at equator of microbubble. (b) Transmission spectra at approximately 780 nm of an OMBR, which was filled with NaCl solution with refractive index difference of 9.5 × 10⁻⁵. Wavelength shift caused by difference in refractive index was 8.56 pm, corresponding to a BRIS of S_bulk = 9 nm/RIU.

Figure 3

Functionalization of OMBR surface and process of specifically identifying HIV-1 p24 antigen.

Figure 4

(a) Fluorescent images of two newly prepared OMBRs after binding fluorescent-labeled IgG. (b) Fluorescent images of antibodies after eight detection cycles with defect points emerging. (c) Fluorescent images of antibodies after hydroxylation of surface. (d) Fluorescent images before (left) and after (right) rinsing with hydrofluoric acid of left OMBR in (c).

Figure 5

(a–e) Wavelength shift curves for detection of 50 fg/mL, 100 fg/mL, 200 fg/mL, 500 fg/mL, and 1 pg/mL (2.1 fM, 4.2 fM, 8.3 fM, 20.8 fM, and 41.7 fM) of HIV-1 p24 antigens, respectively. OMBR was soaked in PBS at the beginning to establish a baseline. Before introduction of antigen, BSA at the same concentration was introduced and did not cause any distinguishable shift. (f) Sensing curve ranged from 50 fg/mL to 1 pg/mL, showing the relationship between wavelength shift and HIV-1 p24 antigen concentration.

Figure 6

(a) Stepped wavelength shifts caused using NaCl solution in linear mass ratio ranging from 0.1% to 0.5%. Before and after NaCl solution was imported into the OMBR, DI water was introduced as a control and as an indication that the system reached equilibrium. (b) Linear fit of wavelength shifts of NaCl solutions. Bulk sensitivity S_bulk = 31 nm/RIU. (c) Here, 1 pg/mL of HIV-1 p24 antigens was detected using three OMBRs with different BRISs in independent experiments. First, 1 pg/mL of BSA was introduced, which did not cause a distinguishable wavelength shift. Then, 1 pg/mL of antigens was introduced at 200 s, and stable dynamic equilibrium was established after 200 s in each detection. Wavelength shifts varied as a consequence of diverse sensitivities. (d) Sensing signals were equalized, and output of three detections was very close regardless of sensitivity differences.

Figure 7

(a) Sensing curves of original repeating method for 15 detections with only one OMBR. Detections were stratified into three series, and in each series, HIV-1 p24 antigens at 50 fg/mL, 100 fg/mL, 200 fg/mL, 500 fg/mL, and 1 pg/mL were detected. (b) Sensing curve of equalization method with 10 OMBRs to detect HIV-1 p24 antigens ranging from 50 fg/mL to 1 pg/mL at intervals of 100 fg/mL. Each OMBR, noted with the same color and shape, detected antigens three to five times and was then replaced with a new one with similar physical characteristics.

Figure 8

(a) Unequalized sensing data ranging from 50 fg/mL to 100 ng/mL in specific detection of HIV-1 p24 antigen. Each OMBR, noted with the same color and shape, detected antigens three to five times and was then replaced with a new one with similar physical characteristics. (b) Equalized sensing data. (c) Relative standard deviation changes before and after equalization. (d) Relationship between S_bulk and wall thickness.